Hi Xtian,
I have added axis lengths as suggested. SVN 561. If
the rank is > 2 then the axis lengths of the higher
dimensions are shown one the same vertical frame line (which I
find less ugly than adding additional
frame lines).
I was wrong when I said that ∈ indicates empty values (it
was long time ago when I implemented that).
Empty values are indicated by ⌽ and ⊖ and the box
then shows the prototype of the empty value.
The ∈ (or several) indicates a nested value (and 8 ⎕CR
automatically encloses its right argument).
There is no such thing as an aggregate in GNU APL - all
values are treated as possibly mixed and/or nested.
Therefore the idea of an integer value or character
value is IMHO a left-over from old APL times where the
data type was in fact a property of the entire value and not a
property of each ravel element. Displaying
such a property in the value frame (which shows only properties of
the entire value) would therefore not
be very plausible.
The new ⎕CR values are 20-24 for different frame
styles.
/// Jürgen
On 03/14/2015 03:50 AM, Christian
Robert wrote:
on
Gnu APL:
8 ⎕cr (3 4 ⍴ ⍳12) (4 5 ⍴ ⍳20)
┌→────────────────────────────┐
│┌→─────────┐ ┌→─────────────┐│
│↓1 2 3 4│ ↓ 1 2 3 4 5││
││5 6 7 8│ │ 6 7 8 9 10││
││9 10 11 12│ │11 12 13 14 15││
│└──────────┘ │16 17 18 19 20││
│ └──────────────┘│
└∊────────────────────────────┘
You said: For example ∈ indicates an empty item in APL2 and not an
aggregate (whatever that may be).
the '∊' means an agregate to me (a mixup of several things)
the fact that it display as '∊' or '∈' is not meaningfull in
itself.
well, it was a suggestion to make it more user friendly as (edited
by hand below)
┌2────────────────────────────┐
│┌4─────────┐ ┌5─────────────┐│
│31 2 3 4│ 4 1 2 3 4 5││
││5 6 7 8│ │ 6 7 8 9 10││
││9 10 11 12│ │11 12 13 14 15││
│└~─────────┘ │16 17 18 19 20││
│ └~─────────────┘2
└∊────────────────────────────┘
an other example:
a←4 4 ⍴ ⍳16
b←⍕a
a b
1 2 3 4 1 2 3 4
5 6 7 8 5 6 7 8
9 10 11 12 9 10 11 12
13 14 15 16 13 14 15 16
8 ⎕cr a b
┌→──────────────────────────┐
│┌→──────────┐ ┌→──────────┐│
│↓ 1 2 3 4│ ↓ 1 2 3 4││
││ 5 6 7 8│ │ 5 6 7 8││
││ 9 10 11 12│ │ 9 10 11 12││
││13 14 15 16│ │13 14 15 16││
│└───────────┘ └───────────┘│
└∊──────────────────────────┘
no thing distinguishes 'a' and 'b'
it should display as:
┌2──────────────────────────┐
│┌4──────────┐ ┌11─────────┐│
│4 1 2 3 4│ 4 1 2 3 4││
││ 5 6 7 8│ │ 5 6 7 8││
││ 9 10 11 12│ │ 9 10 11 12││
││13 14 15 16│ │13 14 15 16││
│└~──────────┘ └───────────┘│
└∊──────────────────────────┘
Xtian.
Xtian.
On 2015-03-13 14:44, Juergen Sauermann wrote:
Hi Xtian,
I will look into this. However a few things seem to conflict
with how GNU APL (and IBM APL2) do it
and I believe this would create too much confusion to be
helpful.
/// Jürgen
For example ∈ indicates an empty item in APL2 and not an
aggregate (whatever that may be).
On 03/13/2015 02:20 AM, Christian Robert wrote:
On my "still" prefered platform
(nars2000) '⎕fmt var' is equivalent to the gnu apl '8 ⎕cr var'
but on that platform ⎕fmt gives more info and this is probably
not hard
to implement also into gnu apl.
here some examples in nars2000
a←⍳10
⎕fmt a
┌10──────────────────┐
│1 2 3 4 5 6 7 8 9 10│
└~───────────────────┘
⍝ the '~' indicate it is an array of numbers, chars
would be the usual '─'
⍝ the '10' is the size
b←3 3 ⍴⍳9
⎕fmt b
┌3────┐
31 2 3│
│4 5 6│
│7 8 9│
└~────┘
⎕fmt a b
┌2─────────────────────────────┐
│┌10──────────────────┐ ┌3────┐│
││1 2 3 4 5 6 7 8 9 10│ 31 2 3││
│└~───────────────────┘ │4 5 6││
│ │7 8 9││
│ └~────┘2
└∊─────────────────────────────┘
⍝ the '∊' indicate an agregate
⍝ the '2' at rightmost is the '≡' (i think)
c←3 3 3 ⍴ 'abcdefghijklmnopqrstuvwxyz0'
⎕fmt c
┌┬3──┐
33abc│
││def│
││ghi│
││ │
││jkl│
││mno│
││pqr│
││ │
││stu│
││vwx│
││yz0│
└┴───┘
⍝ above, it says it is a 3 dimentioanl array of type
'char'
d←3 3 3 ⍴ ⍳27
⎕fmt d
┌┬3───────┐
33 1 2 3│
││ 4 5 6│
││ 7 8 9│
││ │
││10 11 12│
││13 14 15│
││16 17 18│
││ │
││19 20 21│
││22 23 24│
││25 26 27│
└┴~───────┘
⍝ above it says it is a 3 dimentioanl array of type
'numbers'
e←(a b c d)
⎕fmt e
┌4────────────────────────────────────────────────┐
│┌10──────────────────┐ ┌3────┐ ┌┬3──┐ ┌┬3───────┐│
││1 2 3 4 5 6 7 8 9 10│ 31 2 3│ 33abc│ 33 1 2 3││
│└~───────────────────┘ │4 5 6│ ││def│ ││ 4 5 6││
│ │7 8 9│ ││ghi│ ││ 7 8 9││
│ └~────┘ ││ │ ││ ││
│ ││jkl│ ││10 11 12││
│ ││mno│ ││13 14 15││
│ ││pqr│ ││16 17 18││
│ ││ │ ││ ││
│ ││stu│ ││19 20 21││
│ ││vwx│ ││22 23 24││
│ ││yz0│ ││25 26 27││
│ └┴───┘ └┴~───────┘2
└∊────────────────────────────────────────────────┘
⍝ above, an agregate of 4 things
f←e e
⎕fmt f
┌2──────────────────────────────────────────────────────────────────────────────────────────────────────┐
│┌4────────────────────────────────────────────────┐
┌4────────────────────────────────────────────────┐│
││┌10──────────────────┐ ┌3────┐ ┌┬3──┐ ┌┬3───────┐│
│┌10──────────────────┐ ┌3────┐ ┌┬3──┐ ┌┬3───────┐││
│││1 2 3 4 5 6 7 8 9 10│ 31 2 3│ 33abc│ 33 1 2 3││ ││1 2 3 4
5 6 7 8 9 10│ 31 2 3│ 33abc│ 33 1 2 3│││
││└~───────────────────┘ │4 5 6│ ││def│ ││ 4 5 6││
│└~───────────────────┘ │4 5 6│ ││def│ ││ 4 5 6│││
││ │7 8 9│ ││ghi│ ││ 7 8 9││
│ │7 8 9│ ││ghi│ ││ 7 8 9│││
││ └~────┘ ││ │ ││ ││
│ └~────┘ ││ │ ││ │││
││ ││jkl│ ││10 11 12││
│ ││jkl│ ││10 11 12│││
││ ││mno│ ││13 14 15││
│ ││mno│ ││13 14 15│││
││ ││pqr│ ││16 17 18││
│ ││pqr│ ││16 17 18│││
││ ││ │ ││ ││
│ ││ │ ││ │││
││ ││stu│ ││19 20 21││
│ ││stu│ ││19 20 21│││
││ ││vwx│ ││22 23 24││
│ ││vwx│ ││22 23 24│││
││ ││yz0│ ││25 26 27││
│ ││yz0│ ││25 26 27│││
││ └┴───┘ └┴~───────┘2
│ └┴───┘ └┴~───────┘2│
│└∊────────────────────────────────────────────────┘
└∊────────────────────────────────────────────────┘3
└∊──────────────────────────────────────────────────────────────────────────────────────────────────────┘
Thoses informations are particuliarly usefull for everyone
testing/developping
and the actual ]boxing 8 (and 8 ⎕cr) function already know
thoses values (size, rank, types, ...), so they can plug the
info into the *actual* boxes.
I dont mind if you update the actual behavior or define a new
set of ⎕cr function number
or rather define a quite new ⎕fmt to do this. It is quite
missing.
my 2 cents, (would love to see something like that into
gnuapl)
Xtian.
|